Cloud computing offers to users worldwide a low cost on-demand services, according to their requirements. In the recent years, the rapid growth and service quality of cloud computing has made it an attractive technology for different Tech Companies. However with the growing number of data centers resources, high levels of energy cost are being consumed with more carbon emissions in the air. For instance, the Google data center estimation of electric power consumption is equivalent to the energy requirement of a small sized city. Also, even if the virtualization of resources in cloud computing datacenters may reduce the number of physical machines and hardware equipments cost, it is still restrained by energy consumption issue. Energy efficiency has become a major concern for today’s cloud datacenter researchers, with a simultaneous improvement of the cloud service quality and reducing operation cost. This paper analyses and discusses the literature review of works related to the contribution of energy efficiency enhancement in cloud computing datacenters. The main objective is to have the best management of the involved physical machines which host the virtual ones in the cloud datacenters.
Survey: An Optimized Energy Consumption of Resources in Cloud Data Centers
1. Survey: An Optimized Energy Consumption of
Resources in Cloud Data Centers
Sara DIOUANI, Hicham Medromi
Engineering research laboratory
System Architecture Team
ENSEM, HASSAN II University
Casablanca, Morocco
diouanisara19@gmail.com, hmedromi@yahoo.fr
Abstract—Cloud computing offers to users worldwide a low
cost on-demand services, according to their requirements. In the
recent years, the rapid growth and service quality of cloud
computing has made it an attractive technology for different
Tech Companies. However with the growing number of data
centers resources, high levels of energy cost are being consumed
with more carbon emissions in the air. For instance, the Google
data center estimation of electric power consumption is
equivalent to the energy requirement of a small sized city. Also,
even if the virtualization of resources in cloud computing
datacenters may reduce the number of physical machines and
hardware equipments cost, it is still restrained by energy
consumption issue. Energy efficiency has become a major
concern for today’s cloud datacenter researchers, with a
simultaneous improvement of the cloud service quality and
reducing operation cost. This paper analyses and discusses the
literature review of works related to the contribution of energy
efficiency enhancement in cloud computing datacenters. The
main objective is to have the best management of the involved
physical machines which host the virtual ones in the cloud
datacenters.
Keywords—Cloud, green cloud; energy; data center; energy
consumption; virtualization; optimization; resource allocation;
direct migration; consolidation; virtual machine; physical machine.
I. CONTEXT AND ISSUES
Cloud computing is among the important technologies of
the present time. It is modeled to provide services to users [1]
such as computing, software, data access and storage without
any prior knowledge of the physical location and
configuration of the server providing these services. Large and
virtualized data centers contain multiple elements as servers,
networking equipment, cooling systems that consume high
energy to provide efficient and reliable services to their clients
[2].
A report by the International Data Corporation (IDC)
predicts that the space of the Worldwide datacenters will
continue to rise and reach about 1.94 billion square feet in
2018 while it was about 1.58 billion square feet in 2013 [3].
For example, about 70 billion kilowatt-hours of electricity
in 2014 have been consumed by data centers in the United
States, which represents 2 percent of the total country energy
consumption. Moreover, it is expected that the US energy use
will increase by 4% between 2014 and 2020 [4].
The high consumption of energy increases operating costs
for service providers and users. Also, a large amount of carbon
dioxide is emitted which harmfully influences the
environment. On this concern, major research operations on
the energy consumption of the data centers were launched in
the last few years. This problem is not only caused by the
large amount of physical resources, but it resides in using
them in an optimized manner.
Data collected from experiences conducted on more than
5000 servers has shown that server capacity used is between
10 and 50%, which may lead to additional expenses [5].
Besides, the consumption of full capacity on a running server
may reach up to 70% of its power [6].
Virtualization is a very effective high technology that
enables cloud providers to reduce energy inefficiency since it
allows them to manage multiple instances of virtual machines
(VMs) in one server. In fact, by using migration and server
consolidation methods, VMs can be dynamically transferred,
in real time, to a minimum number of physical servers based
on their resources requirements, such as the CPU and
memory. Also, if the server has no VMs, it will be turned off.
These VMs displacements may offer good load balancing
possibilities. It can be done without service disruption so that
it complies with service level agreements (SLA) and the
overall Quality Of virtual Services (QoS). However,
consolidating badly managed VMs can lead to performance
degradation when the demand of resources usage is increased.
Since QoS defined by SLA (such as latency, downtime,
affinity, placement, response time, data duplication, etc) is
largely needed, cloud providers must find a middle ground
between the energy performance of data centers and SLA.
Therefore, effective solutions to manage data centers
International Journal of Computer Science and Information Security (IJCSIS),
Vol. 16, No. 2, February 2018
99 https://sites.google.com/site/ijcsis/
ISSN 1947-5500
2. resources are required to minimize the energy consumption
and maintain a good performance in cloud datacenters
environments. In fact, reducing the energy means reducing
electricity costs, CO2 gas emissions, and contributing to the
green computing aspect.
Green computing is related to several concepts such as:
Power management, efficient algorithms, resources allocation,
virtualization, etc.
Researchers of energy efficiency have proved that a proper
scheduling and management of servers in the cloud
datacenters can efficiently reduce total resources utilization
[7]. The experiment shows that various server components
impact power consumption in the datacenter including CPU,
memory, disk drives, etc [8] as mentioned in the figure 1.
Fig. 1. Range of Power Consumption of Various Server Components
These statistics motivates research efforts in the field of
energy consumption based on multiple substantial energy
parameters.
II. RELATED WORKS
For several years, great efforts have been devoted to the
study of minimizing the energy consumption of data center
servers. Current studies indicate that if this consumption
continues as it is today, the energy cost consumed by a server
during its lifetime will quickly exceed the cost of the server
itself.
Er. Yashi Goyal, et al. [9] presented an approach to select
VM for migration and a host selection policy to reassign this
VM. They rely on the fact that it is better to migrate VMs on
hosts that have low CPU usage so as to not overload and block
the host. As a result, energy consumption is reduced compared
to other approaches.
The below Figure 2 explains live migration; It is the process
of moving a VM or executing applications between different
physical machines PMs without disconnecting the client or
application. The memory and the network connection of VMs
are transferred from the original PM to the destination host.
However, hot-migration or live migration can have a
negative impact on the performance of applications running in
VM [10].
Fig. 2. The concept of live migration
Sandeep Saxena, et al. [11] proposed a task planning
architecture for cloud energy efficiency based on assigning
cloud demand to the most appropriate cloud server.
Noting that, Green Cloud Scheduling [12] refers to the process
of planning cloud service requests in the best way possible in
order to complete the task in time allocated with minimal use
of energy resources. And the energy consumption is increased
if the task has not been properly scheduled. This scheduling
system [2] first classifies incoming cloud service requests into
different categories based on certain predefined parameters
(Bandwidth, Processor, Security and Confidence Level) and
assigns each of the service requests to the cluster that is best
suited to the corresponding category.
Edouard Outin, et al. [13] studied maximization of energy
efficiency by consolidating VM, allocating accurate resources
or adjusting VM usage. They adopted a system that uses
genetic algorithm to optimize energy consumption in cloud
and machine learning technique to improve the fitness
function related to the distributed cluster of the server. As
result, this energy model, can be integrated into the simulator
or other models and it will evaluate the cloud configurations
with more precision. Yet, it does not take into account the
overhead energy costs of live VMs migration.
Chien chich, et al. [14] defined two algorithms; Dynamic
Resources Allocation (DRA) method and Energy Saving
method. With a power distribution unit (PDU) connected to
the system to monitor its status and record energy
consumption. By the DRA algorithm, the waste of the idle
resources on the VMs can be decreased. Also, the Energy
saving method reduces the energy consumption of the cloud
cluster. More precisely, 39.89% of total energy consumption
is reduced (also for memory and VCPUs).
N. Madani, et al. [15][16] developed an architecture for
managing VMs in a data center to optimize energy
consumption by using consolidation (running as many VMs as
possible in a single physical server with avoiding the lack of
resources). This solution considers only the CPU as energy
parameter.
Xin Lu, et al. [17] explained that the modified best fit
decreasing (MBFD) algorithm sorts the VMs in decreasing
order of current usage and allocates each VM to a host that
provides the least increase in power consumption due to this
allocation. However, this sorting algorithm is supposed to
work continuously which generates a large amount of energy
consumption because of the complexity of the algorithm when
implementing a large number of VMs and nodes. The
International Journal of Computer Science and Information Security (IJCSIS),
Vol. 16, No. 2, February 2018
100 https://sites.google.com/site/ijcsis/
ISSN 1947-5500
3. proposed model uses the problem of bin packing (idea to
store items in a minimum number of bins storage, without
exceeding the capacity of bins at all times) [18]; It defines
hotspots hosts in the cloud platform by running the algorithm
selection program. Then, the resources loads of the VMs in
hot spot hosts are ranked in descending order. For non-hot
hosts, their resources loads are sorted in ascending order.
Comparing the model of the scheduling of dynamic migration
of VMs based on MBFD with NPA (Non Power Aware),
DVFS (Dynamic Voltage and Frequency Scaling) and ST
(Simple Threshold) shows that there is more reduction in
power consumption and migration number; 68% and 38% of
energy consumption compared to NPA, DVFS and about 13%
lower for ST [8].
N. Sharma, et al. [19] divided past research on energy
savings into the cloud data center into two main categories
which are: At the single server level and over several servers.
At the single server level, Ch. Wu, et al. [20] used an approach
based on DVFS which takes the task on the basis of its
priorities and minimum order of resources. It generally
reduces the energy consumption of the cloud data center.
It was noticed that the overall use of the data center increases
with the result of lower energy consumption. However, the
lower priority task has a slow response time, thus possibility
of SLA violation.
Beloglazov,et al. [21] proposed a Modified Best Fit
Decreasing (MBFD) algorithm to sort the VMs in descending
order and the PMs in ascending order on the basis of their
processing capacity. After sorting the VMs and PMs, the
distribution of the VMs on the PMs is done by using First Fit
Decreasing (FFD). The limit of this work is that the only
objective is the distribution of VMs. Also, MBFD is not
scalable when a large number of requested VMs arrived at the
data center.
In, An. Xiong, et al. [22], it was shown that other work uses
different algorithms for the allocation of VMs. Including bio-
inspired and nature-inspired algorithms (GA, PSO, OSC, etc.)
for assigning VMs to the cloud data center. Yet, there are
various problems of allocation of VMs with energy efficiency
using PSO. Also, it considers only one type of VM.
Shangg wang, et al . [23] also proposed a Model of
placement of VMs in the data center using PSO with energy
efficiency. Its limitations are that no random reassignment
which is aware of the energy of the VMs after changes of
speed (it takes many iterations and gave a non optimal
solution).
Hadi Goudarzi et al. [24] presented the Constraint
Programming which Consider the VM placement problem to
minimize total energy consumption in a decision time while
maintaining all VMs in the cloud. Multiple copies of VMs are
generated by the approach. The algorithm is designed using
dynamic programming (DP) and local search to evaluate the
number of copies of VM and then place those copies on the
servers in order to minimize the cost of total energy in the
cloud system. The algorithm based on DP: determine the
number of copies of each VM and assign these VMs to the
servers. In the local search method, servers are disabled
depending on their use, and VMs are placed on the rest of the
servers (if possible) to minimize power consumption as much
as possible. This approach reduces the cost of energy by up to
20% compared to previous VM placement techniques.
A. Choudhary, et al. [25] worked on the efficient use of
energy resources of the data center which can be achieved in
two steps. The first one is the efficient placement of VMs and
the second is optimizing resources allocated in the first step by
using live migration as the resources demands change.
The VM Placement aimed to maximize use of available
resources and save energy; As reported by A. Shankar, et
al. [26], the energetic VM placement algorithms include:
Constraint Programming, Bin Packing, Stochastic Integer and
Programming Genetic Algorithm.
And in the context of the live Migration of the VMs for
optimization placement, Anwesha Das [27] described that all
algorithms that attempt to efficiently allocate resources to
demand through live migration answer four questions:
1 Determining when a host is considered overloaded;
2 Determining when a host is considered under- loaded;
3 Selecting the VMs to be migrated from an overloaded
host;
4 Find a new placement of selected VMs for migration
from overloaded and underloaded hosts.
III. SYNTHESIS AND DISCUSION
The table 1 summarizes the most significant aspects of the
notable research in the last seven years, related to the
minimization of energy consumption in datacenter cloud
computing, ordered by year, and mentioning the strategies and
measures (or resources considered: i.e., CPU, memory and so
forth) used, as well as a brief description.
TABLE I. SOME MAIN EXISTING RESEARCHES RELATED TO
MINIMIZATION OF ENERGY CONSUMPTION
References
/ Year
Strategies Measures Description
Mahadevan
et al.[28]
/2010
Consolidating
server load with
network
standby mode
-CPU
-Load link
Server load is
consolidated to fewer
servers and unused
servers and network
items are disabled.
Heller et al.
[29] /2010
Standby mode -Load link Selects a set of active
network elements, then
disables unused links
and switches.
Gmach et
al. [30]/
2010
Dynamic power
management
-Loading
server
Historical traces for
load forecasting,
migration of workloads
from overloaded
servers, shutdown of
slightly loaded servers.
Hsu et al.
[31] / 2012
Virtualization -CPU Limits CPU usage
below the threshold and
consolidate the
workload between
virtual clusters.
Botero et
all. [32] /
2012
Virtual network
integration
-Bandwidth Selects a set of active
network elements, then
disables unused links
and switches.
Xu et al.
[33] / 2012
Green routing -Switch
-Load link
Uses fewer network
devices to provide
routing, inactive
International Journal of Computer Science and Information Security (IJCSIS),
Vol. 16, No. 2, February 2018
101 https://sites.google.com/site/ijcsis/
ISSN 1947-5500
4. References
/ Year
Strategies Measures Description
network devices are
shut down.
Shirayanagi
et al.
[34] /2012
VM
consolidation
with network
standby mode
and bypass
links
-CPU load Combines the placement
of VMs with network
traffic consolidation.
Derivation links are
added to meet the
redundancy
requirements.
Fang et al.
[35] /2012
VM
consolidation,
green routing,
standby mode
-Traffic
rates
Optimizes placement of
VMs and routing traffic
flows through sleep
planning network
elements.
Wang et al.
[23] / 2013
DVFS -Task
execution
time
Non-critical job
execution time is
extended to reduce CPU
voltages.
Beloglazov,
et al. [21]
/2013
MBFD
FFD (First Fit
Decreasing)
-CPU Sort VMs in descending
order and PMs in
ascending order on the
basis of their processing
capacity. After, the
distribution of VMs on
PMs is performed using
FFD.
Damien
Borgetto et
al.[36] /
2014
Constraint
based: SOPVP
-CPU Migrate VM when the
host is under load. The
aim of VM migration is
consolidating servers.
Ajith
Singh. N
latha [37]/
2014
BASIP: banker
algorithm
with SIP
-CPU
-Memory
-Bandwidth
Migrate VM when the
host is overloaded. The
objective of VM
migration is the
attenuation of hot spots.
Zehra
Bagheri, et
al.
[38] / 2014
Bin packing:
Least free
processing
element
-CPU Migrate VM when the
host is overloaded
The goal of VM
migration is
consolidating servers
N. Madani,
et al. [15]
[16]
/2014
VMs
Consolidation
-CPU Architecture for
managing VM in a data
center to optimize
energy consumption by
grafting a component of
multiple consolidation
plans that leads to an
optimal configuration
Ch. Wu, et
al. [20]
/2014
DVFS -CPU This approach takes the
task on the basis of
priorities and the
minimum order of
resources
Er. Yashi
Goyal, et
al. [9]
/2015
Migrate VM on
hosts that have
low processor
utilization
-CPU Approach to select a
VM for migration and a
host selection policy to
reassign this VM
Sandeep
Saxena, et
al. [11]
/2015
Classifies
service requests
into different
categories
based on certain
parameters and
assigns each
request to the
cluster best
-Bandwidth
-Processor
-Level of
security
and trust
Task scheduling
architecture for cloud-
based energy efficiency
based on assigning
cloud demand to the
most appropriate cloud
server.
References
/ Year
Strategies Measures Description
suited to the
corresponding
category
Edouard
Outin, et al.
[13]
/2015
Uses the
genetic
algorithm to
optimize energy
consumption in
the cloud and
learning
techniques
-CPU Maximize energy
efficiency by
consolidating VMs,
allocating specific
resources or adjusting
the use of VMs
Chien
chich, et al.
[14]
/2015
-The Dynamic
Resource
Allocation
(DRA) -The
method of
energy saving.
-Memory
-VCPU
The DRA can reduce
the waste of resources in
the VM and can
increase the allocation
of resources of VMs
with insufficient
resources. The energy
saving method reduces
the energy consumption
of the cloud cluster.
Xin Lu, et
al. [17]
/2015
(MBFD:
modified best
fit decreasing)
-CPU
-Memory
Sorts the VMs in
decreasing order of
current usage and
allocates each VM to a
host that provides less
power consumption
increase due to this
allocation
F. D. Rossi
et al. [39]
/2016
The Energy-
Efficient Cloud
Orchestrator -
e-eco.
-
Transaction
s
per second
-SLA
Decides which energy-
savings technique to
apply during execution
with the cloud load
balancer, to enhance the
trade-off between
energy consumption and
application
performance.
Kejing He,
et al. [40]
/2016
-Improved
Underloaded
Decision (IUD)
algorithm and
Minimum
Average
Utilization
Difference
(MAUD)
algorithm.
-CPU
-SLA
Consolidate VMs using:
in the underloaded
host decision step, the
IUD method that is
based on the overload
threshold
of hosts and the average
utilization of all active
hosts.
And in the step of
selecting the target host
that can accept the vm
migration, the algorithm
MAUD is adopted (that
is based on the average
utilization of the data
center).
M. A.
Khoshkhol
ghi et al.
[41]
/2016
-Energy-
efficient and
SLA-aware VM
consolidation
mechanism.
-CPU
-RAM
-Bandwidth
-SLA
Uses four steps :
-Overloading host
detection, and when
VMs are reallocated to
other hosts.
-Underloading host
detection and when
VMs are consolidated to
other hosts. After,
switching to the sleep
mode the empty host.
-Select the VMs to be
migrated from
overloaded hosts.
-Choose the host for the
International Journal of Computer Science and Information Security (IJCSIS),
Vol. 16, No. 2, February 2018
102 https://sites.google.com/site/ijcsis/
ISSN 1947-5500
5. References
/ Year
Strategies Measures Description
selected VMs.
Riaz Ali, et
al. [42]
/2017
-Energy
efficient VMR
(Virtual
Machine
Replacement)
algorithm.
-Number of
running
physical
servers.
-Turns off idle PMs to
energy saving modes
and then the number of
running PMs is reduced.
Rahul
Yadav, et
al. [43]
/2017
- Power Aware
Best
Fit Decreasing
(PABFD)
algorithm of
VM placement.
-CPU
-SLA
-RAM
Selects VMs to
consolidate from
overloaded or
underloaded host for
migrating them to
another appropriate
host.
Also, the idle hosts are
turned into energy
saving-mode.
What we can analyses from these studies, is that most of the
previous studies do not take into account all the major energy
parameters necessary to ensure the ideal energy efficiency.
In fact, the energy parameters enclose the CPU, the amount of
memory, the disk storage space, the quantity of message
transmitted in the network (bandwidth), the amount of input/
output operations per second (IOPS) available on the physical
support.
Also, the placement of VMs depends on some defined SLA
constraints which may be:
-The affinity constraints between couples of VMs means that
we need to find an optimal placement respecting the fact that
two VMs for example, must be placed on the same physical
server. It is the case of interdependent virtual machines that
share data with each other in short predefined deadlines.
-The security constraints may be for example, separating two
VMs on different servers (or even two data centers).
-The migration constraints may require performing the
placement of VMs exclusively on a set of well-defined PMs,
or even decide to keep a VM on the same server (or even data
center).
We defined other energy parameters including the number of
VMs on the physical machine, the total number of physical
machines used, the number of reallocations of the VM
(displacements), the duration of interruption of a VM in the
migration phase, the percentage of maximum and minimum
consumption of VMs and the response time of a task hosted
by a virtual machine (SLA).
We can also talk about the sustainability of data; each data is
replicated to multiple hosts / data centers in real time (such as
a primary and backup host).
Then, the researches in literature, until that date, are still
lacking and little attention has been given to have a complete
solution enclosing all the major energy parameters and which
covers all possible scenarios and aspects that influence the
energy consumption.
IV. CONCLUSION AND FUTURE WORKS
The field of resources management and energy
consumption is an important and interesting topic in cloud
computing nowadays. In fact, the data centers consume an
enormous amount of electrical energy which causes the
reduction of performances and the emission of a large amount
of carbon dioxide. In order to improve the use of resources
and reduce energy consumption, several technologies are
used, such as server virtualization, migration and
consolidation.
In this paper, we presented an analytical study of the
researches adopted in the literature in the field of the green
cloud to reduce energy consumption of datacenter and achieve
application performance. And as future works, we will
propose a dynamic optimized solution for resources
management through appropriate allocation of VMs in the
cloud data center and which considers most of major energy
parameters and most of possible constraints of VMs allocation
in PMs.
REFERENCES
[1] R. Buyya, C. S. Yeo, S. Venugopal, J. Broberg, and I. Brandic,
“Cloud computing and emerging IT platforms: Vision, hype, and
reality for delivering computing as the 5th utility,” Future Gener.
Comput. Syst., vol. 25, no. 6, pp. 599–616, juin 2009.
[2] R. Sinha, N. Purohit, and H. Diwanji, Power Aware Live Migration
for Data Centers in Cloud using Dynamic Threshold. .
[3] “IDC: Amount of World‟s Data Centers to Start Declining in 2017,”
Data Center Knowledge, 11-Nov-2014. [Online]. Available:
http://www.datacenterknowledge.com/archives/2014/11/11/idc-
amount-of-worlds-data-centers-to-start-declining-in-2017. [Accessed:
26-Jan-2018].
[4] “Shehabi, A., Smith, S.J., Horner, N., Azevedo, I., Brown, R.,
Koomey, J., Masanet, E., Sartor, D., Herrlin, M., Lintner, W. 2016.
„United States Data Center Energy Usage Report‟. Lawrence Berkeley
National Laboratory, Berkeley, California. LBNL-1005775.” .
[5] L. A. Barroso and U. Hölzle, “The Case for Energy-Proportional
Computing,” Computer, vol. 40, no. 12, pp. 33–37, décembre 2007.
[6] A. Beloglazov and R. Buyya, “Adaptive Threshold-based Approach
for Energy-efficient Consolidation of Virtual Machines in Cloud Data
Centers,” in Proceedings of the 8th International Workshop on
Middleware for Grids, Clouds and e-Science, New York, NY, USA,
2010, p. 4:1–4:6.
[7] N. Liu, Z. Dong, and R. Rojas-Cessa, “Task Scheduling and Server
Provisioning for Energy-Efficient Cloud-Computing Data Centers,” in
2013 IEEE 33rd International Conference on Distributed Computing
Systems Workshops, 2013, pp. 226–231.
[8] Y. Sharma, B. Javadi, W. Si, and D. Sun, “Reliability and energy
efficiency in cloud computing systems: Survey and taxonomy,” J.
Netw. Comput. Appl., vol. 74, pp. 66–85, Oct. 2016.
[9] Y. Goyal, M. S. Arya, and S. Nagpal, “Energy efficient hybrid policy
in green cloud computing,” in 2015 International Conference on
Green Computing and Internet of Things (ICGCIoT), 2015, pp. 1065–
1069.
[10] W. Voorsluys, J. Broberg, S. Venugopal, and R. Buyya, “Cost of
Virtual Machine Live Migration in Clouds: A Performance
International Journal of Computer Science and Information Security (IJCSIS),
Vol. 16, No. 2, February 2018
103 https://sites.google.com/site/ijcsis/
ISSN 1947-5500
6. Evaluation,” in Proceedings of the 1st International Conference on
Cloud Computing, Berlin, Heidelberg, 2009, pp. 254–265.
[11] S. Saxena, G. Sanyal, S. Sharma, and S. K. Yadav, “A New Workflow
Model for Energy Efficient Cloud Tasks Scheduling Architecture,” in
2015 Second International Conference on Advances in Computing
and Communication Engineering (ICACCE), 2015, pp. 21–27.
[12] F. Cao and M. M. Zhu, “Energy Efficient Workflow Job Scheduling
for Green Cloud,” in 2013 IEEE International Symposium on Parallel
Distributed Processing, Workshops and Phd Forum, 2013, pp. 2218–
2221.
[13] E. Outin, J. E. Dartois, O. Barais, and J. L. Pazat, “Enhancing Cloud
Energy Models for Optimizing Datacenters Efficiency,” in 2015
International Conference on Cloud and Autonomic Computing
(ICCAC), 2015, pp. 93–100.
[14] C. C. Chen, P. L. Sun, C. T. Yang, J. C. Liu, S. T. Chen, and Z. Y.
Wan, “Implementation of a Cloud Energy Saving System with Virtual
Machine Dynamic Resource Allocation Method Based on
OpenStack,” in 2015 Seventh International Symposium on Parallel
Architectures, Algorithms and Programming (PAAP), 2015, pp. 190–
196.
[15] N. Madani, A. Lebbat, S. Tallal, and H. Medromi, “New cloud
consolidation architecture for electrical energy consumption
management,” in AFRICON, 2013, 2013, pp. 1–3.
[16] N. Madani, A. Lebbat, S. Tallal, and H. Medromi, “Power-aware
Virtual Machines consolidation architecture based on CPU load
scheduling,” in 2014 IEEE/ACS 11th International Conference on
Computer Systems and Applications (AICCSA), 2014, pp. 361–365.
[17] X. Lu and Z. Zhang, “A Virtual Machine Dynamic Migration
Scheduling Model Based on MBFD Algorithm,” Int. J. Comput.
Theory Eng., vol. 7, no. 4, p. 278, 2015.
[18] R. Ren, X. Tang, Y. Li, and W. Cai, “Competitiveness of Dynamic
Bin Packing for Online Cloud Server Allocation,” IEEEACM Trans.
Netw., vol. 25, no. 3, pp. 1324–1331, Jun. 2017.
[19] N. Sharma and R. M. Guddeti, “Multi-Objective Energy Efficient
Virtual Machines Allocation at the Cloud Data Center,” IEEE Trans.
Serv. Comput., vol. PP, no. 99, pp. 1–1, 2016.
[20] C.-M. Wu, R.-S. Chang, and H.-Y. Chan, “A green energy-efficient
scheduling algorithm using the DVFS technique for cloud
datacenters,” Future Gener. Comput. Syst., vol. 37, pp. 141–147,
juillet 2014.
[21] A. Beloglazov and R. Buyya, “Managing Overloaded Hosts for
Dynamic Consolidation of Virtual Machines in Cloud Data Centers
under Quality of Service Constraints,” IEEE Trans. Parallel Distrib.
Syst., vol. 24, no. 7, pp. 1366–1379, Jul. 2013.
[22] A. Xiong, C. Xu, A. Xiong, and C. Xu, “Energy Efficient
Multiresource Allocation of Virtual Machine Based on PSO in Cloud
Data Center, Energy Efficient Multiresource Allocation of Virtual
Machine Based on PSO in Cloud Data Center,” Math. Probl. Eng.
Math. Probl. Eng., vol. 2014, 2014, Jun. 2014.
[23] S. Wang, Z. Liu, Z. Zheng, Q. Sun, and F. Yang, “Particle Swarm
Optimization for Energy-Aware Virtual Machine Placement
Optimization in Virtualized Data Centers,” in Proceedings of the 2013
International Conference on Parallel and Distributed Systems,
Washington, DC, USA, 2013, pp. 102–109.
[24] H. Goudarzi and M. Pedram, “Energy-Efficient Virtual Machine
Replication and Placement in a Cloud Computing System,” in 2012
IEEE 5th International Conference on Cloud Computing (CLOUD),
2012, pp. 750–757.
[25] A. Choudhary, S. Rana, and K. J. Matahai, “A Critical Analysis of
Energy Efficient Virtual Machine Placement Techniques and its
Optimization in a Cloud Computing Environment,” Procedia Comput.
Sci., vol. 78, pp. 132–138, Jan. 2016.
[26] Anjana Shankar, “Dissertation on Virtual Machine Placement in
Computing Clouds; 2010.”
[27] Anwesha Das, “Project dissertation on A Comparative Study of
Server Consolidation Algorithms on a Software Framework in a
Virtualized Environment; 2012.”
[28] P. Mahadevan, P. Sharma, S. Banerjee, and P. Ranganathan, “Energy
Aware Network Operations,” in IEEE INFOCOM Workshops 2009,
2009, pp. 1–6.
[29] B. Heller et al., “ElasticTree: Saving Energy in Data Center
Networks,” in Proceedings of the 7th USENIX Conference on
Networked Systems Design and Implementation, Berkeley, CA, USA,
2010, pp. 17–17.
[30] D. Gmach et al., “Profiling Sustainability of Data Centers,” in
Proceedings of the 2010 IEEE International Symposium on
Sustainable Systems and Technology, 2010, pp. 1–6.
[31] C.-H. Hsu, K. D. Slagter, S.-C. Chen, and Y.-C. Chung, “Optimizing
Energy Consumption with Task Consolidation in Clouds,” Inf. Sci.,
vol. 258, no. Supplement C, pp. 452–462, février 2014.
[32] J. F. Botero, X. Hesselbach, M. Duelli, D. Schlosser, A. Fischer, and
H. de Meer, “Energy Efficient Virtual Network Embedding,” IEEE
Commun. Lett., vol. 16, no. 5, pp. 756–759, mai 2012.
[33] M. Xu, Y. Shang, D. Li, and X. Wang, “Greening data center
networks with throughput-guaranteed power-aware routing,” Comput.
Netw., vol. 57, no. 15, pp. 2880–2899, Oct. 2013.
[34] H. Shirayanagi, H. Yamada, and K. Kono, “Honeyguide: A VM
migration-aware network topology for saving energy consumption in
data center networks,” in 2012 IEEE Symposium on Computers and
Communications (ISCC), 2012, pp. 000460–000467.
[35] W. Fang, X. Liang, S. Li, L. Chiaraviglio, and N. Xiong,
“VMPlanner: Optimizing virtual machine placement and traffic flow
routing to reduce network power costs in cloud data centers,” Comput.
Netw., vol. 57, no. 1, pp. 179–196, Jan. 2013.
[36] D. Borgetto and P. Stolf, “An energy efficient approach to virtual
machines management in cloud computing,” in 2014 IEEE 3rd
International Conference on Cloud Networking (CloudNet), 2014, pp.
229–235.
[37] Ajith Singh. N and M. Hemalatha, “Basip a Virtual Machine
Placement Technique to Reduce Energy Consumption in Cloud Data
Centre - Semantic Scholar,” J. Theor. Appl. Inf. Technol., vol. 59, no.
2, Jan. 2014.
[38] Z. Bagheri and K. Zamanifar, “Enhancing energy efficiency in
resource allocation for real-time cloud services,” in 2014 7th
International Symposium on Telecommunications (IST), 2014, pp.
701–706.
[39] F. D. Rossi, M. G. Xavier, C. A. F. De Rose, R. N. Calheiros, and R.
Buyya, “E-eco: Performance-aware energy-efficient cloud data center
orchestration,” J. Netw. Comput. Appl., vol. 78, pp. 83–96, Jan. 2017.
[40] K. He, Z. Li, D. Deng, and Y. Chen, “Energy-efficient framework for
virtual machine consolidation in cloud data centers,” China Commun.,
vol. 14, no. 10, pp. 192–201, Oct. 2017.
[41] M. A. Khoshkholghi, M. N. Derahman, A. Abdullah, S.
Subramaniam, and M. Othman, “Energy-Efficient Algorithms for
Dynamic Virtual Machine Consolidation in Cloud Data Centers,”
IEEE Access, vol. 5, pp. 10709–10722, 2017.
[42] R. Ali, Y. Shen, X. Huang, J. Zhang, and A. Ali, “VMR: Virtual
Machine Replacement Algorithm for QoS and Energy-Awareness in
Cloud Data Centers,” in 2017 IEEE International Conference on
Computational Science and Engineering (CSE) and IEEE
International Conference on Embedded and Ubiquitous Computing
(EUC), 2017, vol. 2, pp. 230–233.
[43] R. Yadav, W. Zhang, H. Chen, and T. Guo, “MuMs: Energy-Aware
VM Selection Scheme for Cloud Data Center,” in 2017 28th
International Workshop on Database and Expert Systems Applications
(DEXA), 2017, pp. 132–136.
International Journal of Computer Science and Information Security (IJCSIS),
Vol. 16, No. 2, February 2018
104 https://sites.google.com/site/ijcsis/
ISSN 1947-5500